An Enphase Ensemble Installation

svetz

Works in theory! Practice? That's something else
Panels are up, final inspection today hopefully and net-meter should get installed soon. Still about a month away from the batteries. I'll start blogging in this thread all the stuff I learn and a full description of the system as I have time. If there are specific questions you have just post them. You can see pricing information in What should you pay for a Solar Installation?

Here's an introduction to what Enphase Ensemble is:

kWh/month per year
YearJanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecemberMWh/yValue
2020*918129213241348122711981356998875724.7**773**12.2$1610
2021829**928.71303.41300.51410.61194135012481083
* Wetter than normal summer
** Stupid shading Trees
 
Last edited:

svetz

Works in theory! Practice? That's something else
23x LG 340NC1-V5 (340 W)
23x IQ7Plus-72-x-US [240V, 290W]

17 panels facing east ~83°, 6 panels facing west, ~263°.
Panels parallel to roof, about 6" up on top of 2nd story. Roof Slope 6°. Mounted without south tilt to minimize hurricane force wind stresses.

DC Nameplate: 7,820W. From SAM equivalent maximum expected winter: 3,693 W, maximum expected summer: 6,618 W

Location: Key Largo, Florida

Why 290W inverters for 340W panels? Won't that clip?
It's true that the inverters limit the system to 6,670W, but 340W panels only get 340 watts when they meet STC. For this tilt/azimuth the closest they'd come is the summer solstice +/- 30 min from solar noon. SAM does show some clipping, but thanks to the sub-optimal angles it's minor.

If the inverters are only 6.6 kW, why spend extra for 7.8 kW of panels?
The 340W panels capture and generate more power than lower wattage panels even if they're not operating at STC. Consider at 80% light 340W panels generate 272W, whereas 300W panels are only generating 240W.

Why more facing East than West?
On average, local conditions have more later afternoon storms than morning storms, so there's generally more sunlight in the morning. Although the difference isn't that great and I'll expand if the system works well and use the other roof space as well.

Typical Enphase installation, AC off the roof to the Envoy, disconnect, then a tap to the grid. Won't get the Ensemble batteries for a number of weeks, but more on that when it occurs.

Storage: 13.44 kWh, 5.12 KVA continuous, 7.68 KVA for 10s, 92.8 amps for 3 cycles... Encharge 3 & Encharge 10

Monitoring
: Home brew program uses the Envoy Api and an reading the Net Meter via RF (see also Off-grid Solar / Battery monitoring and control freeware), and Foghorn, a battery alert system for when off-grid.

Automation: Working on it, possibly FireStick based
 
Last edited:

Picasso

Solar Addict
Super happy with my LG panels, outperform my Solarworld, REC, and Canadian solar panels in shading and cloudy cover. If I could find them near me for under $1 watt I'd love more. Also a HAPPY enphase investor......
 

svetz

Works in theory! Practice? That's something else
Panels and Racking
Salt, wind, and Pmax where my big considerations in selecting panels (see Comparing Solar Panels) and my two top choices were the Panasonic HIT or the LG Neon. Both are great panels from what I can tell. It looked like the Panasonic's higher voltage gave it an efficiency edge and it was my first choice, but the LG's had a hefty price break. SAM simulations for the 25 years were about neck and neck for degradation, but Panasonic seems to start pulling ahead after year 25.

I had expected the standard two-rail installation, much to my surprise the installer did 3 rails and each panel has 6 clamps points. I'll get some photos up as I have time.

As you'd expect, the microinverters are attached to the rails and reside under the solar panel in their shade. The AC cables are daisy-chained from panel to panel and each of the rails is grounded to the next and carried down to the whole-house ground.

No Tilt?
Yes, it's a 7.5% loss over the optimum tilt according to SAM and they don't self-clean properly. At least flat is best for summer when I use AC, right? Not really, summer is also the stormiest so with Florida's great net-metering deal I'd do better optimizing for better weather.

Flat is because of roof-strength. I have a great roof, but, Force (N) = area in m² ×(1.229 kg/m³) × (air-speed in m/s)². Structures in my High-Velocity Hurricane Zone (HVHZ) are supposed to survive 180 mph winds. That's a dynamic pressure of ~0.57 psi, doesn't sound like much except
my panels are 2656 in² making a total force of 1500 lbf per panel if perpendicular. From trig, even a small tilt angle of 5° transfers about 1% of the force downwards, so multiply that by the number of panels and it's a lot of overall downwards force.

Ironridge and Unirac both have a nice calculators that you can figure out what you need in the way of attachments.
 
Last edited:

Picasso

Solar Addict
The major plus on the panasonics for me is the little rain channel, sure I could jig and route one but thats more work....Ive yet to see that little rain channel in any other mfg. if your gasketing between panels for a usable space under thepanels the little channel helps.
 

svetz

Works in theory! Practice? That's something else
The power company didn't get queued up, but theoretically they are now and they're "fast", by which they mean less than a week to get the net-meter installed.
 

svetz

Works in theory! Practice? That's something else
Svetz, you putz, you should have done this last year and saved money!

No duh. With this much money the difference between a 30% tax break and 26% is noticeable. But I wanted to get some things done to the roof and then I wanted to get the roof re-coated. In typical contractor style for here, those tasks took insanely long. Still, when I made the deal with my solar installer they promised they'd get it done before the end of the year. That didn't happen, but after a discussion with them, they agreed to cover the difference. Whew!
 

svetz

Works in theory! Practice? That's something else
This is before the net-meter installation, so the panels are in the sun but not exporting power.

Capture.PNG
 

svetz

Works in theory! Practice? That's something else
The net-meter was installed, pity I didn't film it for you....

Whenever I've seen a meter installed they pulled the energized side out first and going in it is last. But this guy was putting in the energized side first. His supervisor was standing right there and didn't say anything so I figured I was learning something new about net-meters.

Yeah... Energized side went in first, but the meter wasn't square and the leg touched the outside of the can. BOOM! Sparks everywhere.

It didn't weld itself and the installer was cool as a cucumber. He just pulled it out, and then put it in again using the same technique. Then thumped it good with his palm and put on on the tags. I was a little surprised the meter wasn't fried.
 

svetz

Works in theory! Practice? That's something else
Jay2020 posted this in the dirty electronics thread...

The radiation occurs due to the high currents ... To fix this problem ... twist the positive and negative battery cables together as tightly as possible...
This explains something I saw the installer do and didn't understand....they twisted the DC lines from the panels to the microinverters (the AC lines aren't twisted). I don't have a good photo of it, but you can see it in this photo. Makes great sense now thanks to Jay2020!

Capture.PNG
 

svetz

Works in theory! Practice? That's something else
I thought you liked Sol-Ark?
I still do, even more so with there new offering <sigh>. The problem for me is the basic flood elevation (BFE) here is 9'. Since they don't come in a NEMA enclosure they need to go inside, which would mean upstairs in living space; for me that's not going to happen. The big advantage of Enphase microinverters is that they're on the roof, and all the other equipment NEMA 3r so is bolted to an outside wall fairly high up.

Well, it all worked out then...right?
Not quite. The battery Enphase has is slightly above the price point of the Kilovault. I'm sure, like the kilovault, that it is a quality product. But, I really wish I had the opportunity to add onto it DIY style as it would be fun and I could take advantage of the great new products Will keeps finding. I haven't yet seen how I can add AC batteries in the future (see Incrementally Adding AC Batteries) and I'm not aware of anyway to expand the Enphase system other than buying more Enphase gear (and I hate being locked into a vendor).
 
Last edited:

Picasso

Solar Addict
well void the warranty and crack open an ensemble battery.....swap out the life4 cells for bigger. People hacked on the first and 2nd gens..people will play with the new ones im sure.
 

svetz

Works in theory! Practice? That's something else
I'd rather add to than swap out, but my current plan is see what I can figure out once it's here (preferably without voiding the warranty if possible). I know they have an expansion plan as you can parallel the devices, but I'm not sure what they're doing electronically. But I'll post what I learn.

As AC batteries become more commonplace with solar I expect somebody (like Outback) will probably come up with a generic solution for the residential marketplace (from Incrementally Adding AC Batteries we know there are solutions for largish commercial operations) to compete with things like Enphase's expansion battery.
 
Last edited:

svetz

Works in theory! Practice? That's something else
Monitoring
So there's a free web page that Enphase has that let's you see what your panels are doing (ref). That's nice, I can even see from 11:30ish to 1:00ish one of the panels has some shade from the power lagging behind the others (confirmed with visual observation).

But really I'd like a program that monitors each panel and lets me know if it's time to go up and clean them or if something bad happen (e.g., not producing, unexpected shade). They have a pay-for monitoring service, but I'm not sure how good it is or if its worth it. I'll find out more about that later.

When I read the service agreement for the free web page (I was curious what they'd be doing with my data) I learned they have an API from which you could make 10k hits/month for free. See https://developer.enphase.com/docs.

Easy enough to calculate the maximum power based on sun and panel orientation for any time of the day. Pretty easy to get the panel data from the API. Pretty easy to get local weather conditions via openweathermap or an IP enabled weather station if you have one. Pretty easy to store historical data and know what a panel's production ranges are.

The service meter is read by RF. Could I use that? If so I'd know how much power I was putting onto the grid and how much I was consuming from the grid. Which means I could know what the house was consuming at any time. Turns out the answer is yes: https://hackaday.com/2014/02/25/using-sdr-to-read-your-smart-meter/. Basically using a $20 RTL-SDR dongle you can read the meter. Douglas Hall wrote a program to decode and report the information, so even easier! The raw data looks something like:

meters_at_my_house.png

So this might be a fun project!

Update: You can get weather direct to if you live near the coast (refA, refB, probably inland services too):
NavTex is a marine digital data radio service designed for transmitting information like navigational and meteorological warnings, weather forecasts and maritime safety information. It is broadcast in either the MW frequency band at 490 kHz and 518 kHz or in the HF band at 4209.5 kHz.

Update 2: Bought a RTL-SDR dongle and will play with it - but I'll blog that in another thread.
 
Last edited:

svetz

Works in theory! Practice? That's something else
APIs for Monitoring

In addition to what's mentioned above, the Envoy has an API. This seems a lot more useful than the developer API. Here's what I've learned so far...
They also have a restful api, some of the commands require a digest username password which is "envoy" and the last 6 digits of the serial number. Here's what I know so far:
{​
"wattHoursToday": 2073,​
"wattHoursSevenDays": 140689,​
"wattHoursLifetime": 144796,​
"wattsNow": 1362​
}​

{​
"serialNumber": "<redacted>",​
"lastReportDate": 1580567783,​
"devType": 1,​
"lastReportWatts": 58,​
"maxReportWatts": 247​
},​

Obviously there must be a whole lot more, I'll see what I can find.

Updates:
 
Last edited:

svetz

Works in theory! Practice? That's something else
The RF dongle arrived on Monday, but didn't get time to play with it until today. Got it to read the utility meter fairly easily, see Using RF to read ITRON utility meters for how that went, one more puzzle piece solved for when I get around to writing monitoring software. The dongle gets pretty warm while it's reading RF. Not so hot you have to pull your fingers away, but fairly close; probably won't be a problem.
 

svetz

Works in theory! Practice? That's something else
Ensemble Battery Specs

Dimensions (WxHxD) 14.45 in x 26.14 in x 12.56 in42.13 in x 26.14 in x 12.56 in
Rated (continuous) output power1.28 KVA3.84 KVA
Peak output power1.92 kVA (10 seconds)5.7 kVA (10 seconds)
Rated Output Current5.3 A16 A
Usable capacity3.36 kWh10.08 kWh
Round trip efficiency96%96%
Optimum operating temperature range0º C to 30º C (32º F to 86º F)0º C to 30º C (32º F to 86º F)
CoolingNatural convection – No fansNatural convection – No fans
ChemistryLithium iron phosphate (LFP)Lithium iron phosphate (LFP)
Enclosure NEMA-3RNEMA-3R
CompatibilityGrid-tied PV systems & EnphaseGrid-tied PV systems &Enphase
Limited Warranty>70% capacity for 10 years or 4000 cycles>70% capacity for 10 years or 4000 cycles

The last one (compatibility) is interesting, sounds like it'll work with non-Enphase GT PV systems. Labels at the top are links to the datasheets.
 
Last edited:
Top